Papers
arxiv:2611.00871

ShiftRAG: Bypassing the Textual Bottleneck via Decoupled Learning and Continuous Soft Tokens

Published on Oct 31
Authors:
,
,

Abstract

A decoupled retrieval-augmented generation framework uses soft-token interfaces and orthogonality objectives to improve retrieval and question answering while preserving the language model's general capabilities.

Despite advances in retrieval-augmented generation (RAG), suppressing hallucinations during complex reasoning remains a persistent challenge. We identify that existing RAG pipelines suffer from a textual bottleneck: by passing only raw text to the generator, they discard discriminative signals of the retriever. Furthermore, recent attempts to unify retrieval and generation within a single LLM induce severe objective conflict and exacerbate the rank collapse of decoder-only architectures, degrading performance on both tasks. To address these limitations, we introduce ShiftRAG, a decoupled RAG framework that employs a single LLM backbone but isolates discriminative retrieval learning from autoregressive generation. To bridge these decoupled modes, we project retrieval alignments into continuous soft tokens, establishing a high-bandwidth interface that infuses the generator with retrieval-side signals. Concurrently, our soft orthogonality objective mitigates rank collapse, while layer-wise relevance dynamics preserve the fine-grained latent space required for highly discriminative evidence separation. Extensive evaluations demonstrate that ShiftRAG significantly improves retrieval and question answering performance over state-of-the-art baselines. ShiftRAG preserves foundational capabilities of the backbone while maintaining a favorable efficiency-performance trade-off on general text encoding and generative tasks.

Community

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2611.00871
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2611.00871 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2611.00871 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2611.00871 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.